PDF rendering of images seems to vary from viewer to viewer with blurring?

Question

I'm generating PDF figures in Adobe Illustrator CS5 that include embedded images. I've noticed that the images look dramatically different when I display the same PDF in Preview, Skim or Adobe Reader (I'm on OS X). See screenshots. Adobe Reader displays them "correctly" while Skim and Preview blurs the image out each in a different way.

Is there a setting I can set when saving my PDF from Illustrator so that the images are displayed correctly in Skim and Preview?

The PDF was generated in Illustrator and saved without any compression or downsampling.

The original PDF is here: http://ge.tt/8iZMR2A

Adobe Reader 9

Skim 1.3.18

Preview 4.2

Super User's client-side PDF renderer

Answer 1

Most likely both Skim & Preview are buggy in that they don't respect Interpolate flag on the image. We used to have a similar bug in SumatraPDF.

If that's the case, unfortunately there's nothing you can do.

You can try to set it explicitly to false (it's supposed to be false by default, but it's possible that Skim & Preview assume it's true and would work correctly if it's set explicitly to false) but I don't know how you can do that in Illustrator (or if it even allows for such level control).

You might wonder: what is this Interpolate flag I speak of? Unfortunately, explaining this requires explaining PDF file format, which is beyond the scope of this answer. The flag is described in page 207 of the PDF spec (currently linked from here).

Answer 2

Different viewers use different viewing implementations based on their interpretation of the PDF spec version(s) they are aiming to support.

This will be further compounded when you start crossing platforms, which can use entirely different graphic rendering subsystems (etc.).

In much the same way as there's no "Make this look exactly the same on every browser" option in DreamWeaver, I think you're best best is to figure out a set of settings that make your PDFs look as good as possible on as many different readers as you can. That will take testing and tweaking to develop repeatable procedures that can be used on future PDFs.